Abstract
Background/Aim: This study evaluated the efficacy of adjuvant chemotherapy (AC) for intraductal papillary mucinous carcinoma (IPMC). Patients and Methods: We retrospectively analyzed patients who underwent pancreatectomy for invasive IPMC from January 2007 to June 2020. We evaluated outcomes of AC in the entire cohort and in patients with known prognostic factors. Results: A total of 51 patients with invasive IPMC underwent surgery, of which 35 received AC. In the entire cohort, there was no significant difference in median overall survival (OS) between the AC and surgery alone (SA) group [hazard ratio (HR)=0.54; p=0.232]. For patients with poorly differentiated adenocarcinoma, median OS was significantly longer in the AC group (HR=0.27; p=0.022). For patients with lymph node metastasis, median OS was significantly higher in the AC group (HR=0.07; p<0.001). Conclusion: AC may be effective for selected invasive IPMC patients.
- Adjuvant chemotherapy
- invasive IPMC
- poorly differentiated adenocarcinoma
- lymph node metastasis
- tubular type
Intraductal papillary mucinous neoplasms (IPMNs) of the pancreas are cystic neoplasms characterized by mucin-producing epithelial cells lining the pancreatic ducts, forming papillary structures and dilated cystic ducts (1). IPMNs are categorized as: low-grade dysplasia, high-grade dysplasia, or invasive intraductal papillary mucinous carcinoma (IPMC) based on the degree of cellular atypia and the growth pattern of the lining epithelium (2). As the term “minimally invasive” IPMC has been inconsistently defined, it is preferable to stage invasive IPMC using conventional American Joint Committee on Cancer/tumor-node-metastasis classification (3).
Surgery is the standard of care for invasive IPMC (3). The post-resection 5-year overall survival (OS) rate of invasive IPMC ranges from 22% to 62%, which is significantly better than that of pancreatic ductal adenocarcinoma, which ranges from 10% to 20% (4-10). Invasive IPMC are conventionally classified into three histological subtypes: the tubular, colloid, and oncocytic types. The tubular type has poor OS, which is comparable to pancreatic ductal adenocarcinoma (PDAC), while a better prognosis is reported for the colloid and oncocytic types (11, 12).
Adjuvant chemotherapy (AC) has been shown to prolong OS in resected pancreatic cancer (13). In Japan, tegafur/gimeracil/oteracil (S-1) is the standard AC for pancreatic cancer based on the results of the JASPAC 01 study (14). However, only 5% of patients included in this study had invasive IPMC. Therefore, the effectiveness of AC for resected invasive IPMC remains unclear (7, 9, 15–23). While effectiveness in the tubular type has been reported (17), multiple subtypes of invasive IPMC are often observed in the same specimen. In clinical practice, it is often difficult to determine the dominant subtype because there are no clear diagnostic criteria for subtypes of invasive IPMC. Effectiveness of AC in patients with poorly differentiated adenocarcinoma (7, 9) or lymph node metastasis (9, 17, 22) has also been reported. Reports evaluating the efficacy of adjuvant S-1 for resected invasive IPMC are scarce.
In this study, we aimed to evaluate the efficacy of AC, mainly with S-1, for patients with resected invasive IPMC.
Patients and Methods
Study design. We conducted a single-center retrospective cohort study of patients with invasive IPMC who underwent pancreatic resection at the Cancer Institute Hospital of Japanese Foundation for Cancer Research. This study was approved by the Ethics Committee of Cancer Institute Hospital of Japanese Foundation for Cancer Research (No: 2021-1034) and performed in accordance with the Declaration of Helsinki. Since all data were collected from a medical record system, the Ethics Committee waived the requirement to obtain individual informed consent.
Patient selection. We reviewed our database for patients who underwent surgical resection for invasive IPMC from January 2007 to June 2020. The inclusion criteria were as follows: 1) histologically-proven invasive component of adenocarcinoma arising from IPMN, 2) pathological stage I, II, or III according to the TNM Classification based on the AJCC/UICC staging for exocrine pancreatic cancer 8th edition 24, and 3) no local residual tumor (R0) or microscopic residual tumor (R1) after surgery. Patients were excluded from the study if they: 1) received AC at another hospital, 2) had concurrent PDAC, 3) received neoadjuvant therapy, 4) had cancer cells in the intraoperative peritoneal lavage fluid cytological examination, 5) died in the post-operative period (within 30 days), or 6) had distant metastases when AC was commenced.
Clinical data collection. Clinical variables included patient age, sex, and performance status (PS) as per the Eastern Cooperative Oncology Group (25). Performance status was determined from medical records of follow-up visits on the day AC was introduced or, for patients who did not undergo AC, three months after surgery. We also collected information on the type of pancreatectomy, surgical operation time, blood loss during surgery, concomitant vascular resection rate, and the Clavien-Dindo classification (26).
Pathological definition and data collection. Resected specimens were fixed with 20% buffered formalin. Serial 5 mm sections were prepared, with axial slicing for the pancreatic head and sagittal slicing for the pancreatic body. Each section was embedded in paraffin and 3 μm slices were prepared for staining with hematoxylin and eosin. All available stained sections with an invasive component were re-reviewed by a single pathologist (MT). Tumor size was defined as the largest dimension of the invasive component. Resection margin was defined as follows: no local residual tumor (R0), microscopic residual tumor (R1), and macroscopic residual tumor (R2). Microscopic perineural and lymphovascular invasion were also evaluated.
Invasive IPMC was defined as invasive adenocarcinoma arising from non-invasive mucinous epithelial neoplasm in pancreatic ducts. In addition, since PDAC sometimes forms dilated pancreatic ducts, which resemble IPMN, we only included cases in which non-invasive papillary components could be recognized within a dilated pancreatic duct. The histological subtypes of invasive components were classified as either tubular, mucinous, or oncocytic (Figure 1A-C). The definitions of invasive subtypes were as follows: the tubular type, invasive components predominantly consisting of tubular glands; the mucinous type, invasive components producing intraglandular mucin or forming mucin-lake; the oncocytic type, invasive components predominantly consisting of tumor cells with remarkably eosinophilic cytoplasms. When more than two subtypes coexisted, the invasive component was considered to be the most dominant subtype. The histological grade of the invasive component was classified into two groups: 1) poorly differentiated adenocarcinoma and 2) well or moderately differentiated adenocarcinoma (Figure 1D-F). Adenocarcinomas were considered to be poorly differentiated if any poorly differentiated carcinoma cells were identified within the invasive component. We also calculated the proportion of carcinoma cells within the invasive component, which were poorly differentiated.
Histological subtypes and grades of invasive components in resected IPMC. (A) Tubular type. Scale bar: 300 μm. (B) Mucinous type. Scale bar: 300 μm. (C) Oncocytic type. Scale bar: 300 μm. (D) Well differentiated adenocarcinoma. Scale bar: 300 μm. (E) Moderately differentiated adenocarcinoma. Scale bar: 300 μm. (F) Poorly differentiated adenocarcinoma. Scale bar: 300 μm. IPMC: Intraductal papillary mucinous carcinoma.
Adjuvant chemotherapy. We reviewed medical data of 51 patients with resected invasive IPMC. Thirty-five patients received AC. Thirty-two patients received oral S-1 monotherapy twice per day, receiving 40 mg per dose if the body surface area was less than 1.25 m2, 50 mg per dose if the body surface area was 1.25 m2-1.50 m2, or 60 mg if the body surface area was 1.5 m2 or more. S-1 monotherapy was either given as a 6-week regimen (administration for 28 consecutive days followed by a 14-day rest, for up to four cycles) or as a 3-week regimen (administration for 14 consecutive days followed by a 7-day rest, for up to eight cycles). The choice of regimen was left to the physicians’ discretion. Three patients received gemcitabine monotherapy, consisting of intravenous gemcitabine at a dose of 1,000 mg/m2 over 30 min on days 1, 8, and 15 every 4 weeks, which was repeated for up to six cycles. Relative dose intensity (RDI) was calculated as the ratio of the actual dose intensity to the scheduled dose intensity. The actual dose intensity was calculated as the ratio of the actual dose to the actual duration of chemotherapy, while the scheduled dose intensity was calculated as the ratio of the scheduled dose to the scheduled duration of chemotherapy. Patients were included in this study even if cancer recurrence was observed before completing the planned AC schedule.
Outcomes. Patients who received AC and those who underwent surgery alone (SA) in the entire cohort were compared. The primary outcomes were 3-year and 5-year survival rates and OS. The 3-year and 5-year survival rates were defined as the percentage of patients who were alive and followed up 3 and 5 years after surgery, respectively. OS was defined as time from surgery to the date of death from any cause or the last follow-up. We also evaluated 3-year and 5-year recurrence-free survival (RFS) rates, which was defined as time from surgery to the date of first recurrence in the residual pancreas or in other organs including distant lymph nodes noted on imaging studies, death from any cause, or the last follow-up. The effectiveness of AC on invasive IPMC with three reported prognostic factors (poorly differentiated adenocarcinoma, lymph node metastasis, and tubular subtype) were also analyzed. Adverse events were assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 5.0) (27).
Statistical analysis. Categorical variables are expressed as percentages, while quantitative variables are expressed as medians or means. The chi-square test and Fisher’s exact test were used to evaluate statistical differences in categorical variables, as appropriate. The Mann-Whitney U-test was used to examine differences in consecutive non-parametric variables, while the Student’s t-test was used for consecutive parametric variables. Survival analyses were conducted using the Kaplan-Meier method and differences in survival were compared using the log-rank test. A p-value <0.05 was considered statistically significant. Statistical analyses were carried out using EZR software version 1.40 (28).
Results
A total of 51 patients with invasive IPMC were included in the analysis (Figure 2). Thirty-five patients were in the AC group. Sixteen patients were in the SA group. Among patients in the SA group, eleven did not receive AC due to the physician’s discretion in the era before the results of the JASPAC 01 study were revealed, and five did not due to postoperative complications. AC was introduced after a median interval of 8 weeks after surgery, and within 12 weeks after surgery in 82% of patients. Median RDI in the AC group was 95% (range=21-100%). The median follow-up period was 38.9 months (range=20.1-83.5 months) and 30.8 months (range=11.9-130.4 months) for patients who were still alive in the AC and SA groups, respectively. During the follow-up period, cancer recurrence was observed in 19 patients (pancreas: 2, liver: 6, lung: 7, ovary: 1, peritoneum: 5, lymph nodes: 8).
Study flow diagram. IPMN: Intraductal papillary mucinous neoplasm; IPMC: intraductal papillary mucinous carcinoma; AC: adjuvant chemotherapy; PDAC: pancreatic ductal adenocarcinoma.
Adjuvant chemotherapy. Patient characteristics of the AC and SA groups are shown in Table I. Patients in the AC group had significantly better PS. Lymphovascular and perineural invasion and lymph node metastasis were significantly more common in the AC group. There was also a significant difference in the distribution of invasive subtypes.
Characteristics of patients receiving adjuvant chemotherapy versus surgery alone.
Patients in the AC and SA groups demonstrated 5-year survival rates of 54.9% and 55.2%, respectively (p=0.226) (Table II). Median OS was not reached in either group [hazard ratio (HR)=0.54; 95% confidence interval (CI)=0.20-1.48; p=0.232] (Figure 3A and Figure 4A). Five-year RFS rates were 51.0% and 47.3%, respectively (p=0.368) (Table II). Median RFS was not reached in the AC group and 23.4 months in the SA group (HR=0.66; 95%CI=0.27-1.63; p=0.372) (Figure 4B).
Uni-and multi-variate analyses to identify clinicopathological characteristics related to relapse-free survival (n=52).
Kaplan-Meier survival curves depicting (A) OS in the AC and SA groups for the entire cohort. (B) OS in the AC and SA groups for patients with poorly differentiated adenocarcinoma. (C) OS in the AC and SA groups for patients with lymph node metastases. (D) OS in the AC and SA groups for patients with the tubular type. OS: Overall survival; AC: adjuvant chemotherapy; SA: surgery alone.
Forest plots of the impact of AC on overall survival (A) and recurrence-free survival (B) in prespecified subgroups. AC: Adjuvant chemotherapy; SA: surgery alone; N+: lymph node metastasis; N0: no lymph node metastasis: CI: confidence interval.
Of note, the slope of the OS curve in the SA group became flatter after 20 months. There were four AC patients who had an OS of less than 20 months. Their median age was 81.5 years (range=73-86 years) and PS was 3 in two of the four patients. On the contrary, the OS curves of the two groups overlapped after 40 months. There were four SA patients who achieved an OS exceeding 40 months. Invasive components of IPMC in three of these patients had purely well differentiated adenocarcinoma. The invasive component in the other patient was 95% well differentiated and 5% poorly differentiated adenocarcinoma. Resected lymph nodes were negative in all four patients.
Poorly differentiated adenocarcinoma. Thirty-four patients had invasive IPMC with poorly differentiated adenocarcinoma (Supplementary Table I). Patients with poorly differentiated adenocarcinoma in the AC group were significantly younger and had better PS than those in the SA group. There was also a significant difference in the distribution of surgical complications based on the Clavien-Dindo classification.
Patients in the AC group demonstrated a better 3-year survival rate than in the SA group, with 64.4% and 26.7% (p=0.015) (Table II). Median OS was not reached in the AC group and 30.4 months in the SA group (HR=0.27; 95%CI=0.09-0.83; p=0.022) (Figure 3B and Figure 4A). Patients in the AC and SA groups had 3-year RFS rates of 54.2% and 27.8%, respectively (p=0.105) (Table II). Median RFS was not reached in the AC group and 23.4 months in the SA group (HR=0.42; 95%CI=0.14-1.24; p=0.116) (Figure 4B).
Lymph node metastasis. Twenty-seven patients had lymph node metastases (Supplementary Table II). Patients in the AC group were significantly younger and had better PS compared to that in the SA group.
Patients in the AC group demonstrated a better 5-year survival rate compared to that in the SA group, with 50.1% and 0%, respectively (p<0.001) (Table II). Median OS was not reached in the AC group and 8.4 months in the SA group (HR=0.07; 95% CI=0.02-0.28; p<0.001) (Figure 3C and Figure 4A). The AC group also had a better 5-year RFS rate compared to that in the SA group, 44.9% and 0%, respectively (p<0.001) (Table II). Median RFS was 44.4 months and 4.2 months in the AC and SA group, respectively (HR=0.14; 95%CI=0.04-0.44; p<0.001) (Figure 4B).
Tubular type. Thirty-nine patients had tubular invasive IPMC (Supplementary Table III). Patients in the AC group were significantly younger and had better PS compared to that in the SA group.
Patients in the AC group and SA group had 5-year survival rates of 51.3% and 35.6%, respectively (p=0.047) (Table II). Median OS was not reached in the AC group and 30.4 months in the SA group (HR=0.35; 95%CI=0.12-1.03; p=0.057) (Figure 3D and Figure 4A). Patients in the AC and SA groups had 5-year RFS rates of 47.9% and 35.6%, respectively (p=0.251) (Table II). Median RFS was 44.4 and 23.4 months in the AC and SA groups, respectively (HR=0.55; 95%CI=0.20-1.55; p=0.259) (Figure 4B).
Prognostic factors. The results of univariate Cox regression analyses for OS and RFS are shown in Table III. AC was not significantly associated with either improved OS or RFS. Poorly differentiated adenocarcinoma, lymph node metastasis, and tubular type showed tendency toward shorter OS (HR=2.64, 2.81, and 3.58, respectively) or RFS (1.76, 2.46, and 2.08, respectively); however, they were not statistically significant. Older age and decreased PS were associated with shorter OS (HR=1.11 and 2.62, respectively) and shorter RFS (HR=1.06 and 1.94, respectively).
Univariate analysis of factors associated with overall survival and recurrence-free survival.
Adverse events. Grade 3 neutropenia, anemia, and fatigue were observed in patients who received S-1 monotherapy. No grade 4 or 5 adverse events were observed in patients who received AC (Table IV).
Adverse events in patients who received adjuvant chemotherapy.
Discussion
In this study, we evaluated the effectiveness of AC, mainly with S-1, after surgery for invasive IPMC. However, AC did not improve OS or RFS for the entire cohort, similar to previous reports that mainly evaluated the efficacy of 5-fluorouracil or gemcitabine-based adjuvant therapy (15, 17-21). On the contrary, AC may benefit patients with poorly differentiated adenocarcinoma or lymph node metastases.
The OS curves may suggest that some patients did not tolerate AC well, whereas others did not require AC to remain recurrence-free. The four AC patients with a short OS may not have tolerated AC due to their older age and decreased PS. On the contrary, long-term OS curves of the two groups overlap beyond 40 months, with the four remaining patients in the SA group having well differentiated adenocarcinoma with no lymph node metastasis. Patients with these characteristics may not require AC.
Based on the results of univariate Cox regression analysis in this study, poorly differentiated adenocarcinoma and lymph node metastasis showed a tendency toward shorter OS or RFS; however, they were not statistically significant. The small sample size may be the reason for these results. In fact, patients with poorly differentiated adenocarcinoma and lymph node metastases in the AC group showed better survival compared to those in the SA group, with HRs of 0.27 and 0.07, respectively. Some previous studies have also reported favorable results from AC in patients with poorly differentiated adenocarcinoma and lymph node metastasis. McMillan et al. (9) found that AC led to a longer median OS of 15.5 months in patients with poorly differentiated adenocarcinoma compared to 9.6 months in those undergoing SA. They also found that AC improved OS in patients with lymph node metastasis, with a median OS of 19.4 months compared to 12.2 months in the SA group. However, it should be noted that this study included some patients who underwent chemoradiotherapy. Marchegiani et al. (17) also reported that AC improved OS in patients with lymph node metastasis. The rate of 5-year disease-specific survival in the AC group was 76% compared to 35.8% in the SA group. However, AC was not associated with improved OS in poorly differentiated adenocarcinoma patients, with a 5-year disease-specific survival rate of 32% compared to 27.2% in the SA group. However, in this study, patients received adjuvant therapy based on various treatment protocols. In contrast, more than 90% of patients in our study received AC with S-1 based on the treatment protocol of the JASPAC 01 trial (14).
Our study also suggests potential benefit from AC in patients with tubular type invasive IPMC. Patients in the AC group had a tendency towards better survival compared to the SA group. One previous study reported that AC was associated with a significantly improved 5-year disease-specific survival rate in patients with the tubular type (89% vs. 53%) (20). However, this study excluded all cases with mixed subtypes from its analysis. In our study, we conducted analyses based on the most dominant subtype. One-third of invasive IPMCs with the tubular type contained mucinous or oncocytic components, which have been associated with better survival in comparison to the tubular type. Our definition of invasive subtypes may explain why AC was not significantly better than SA in this subgroup. We believe our results are more clinically relevant, as mixed subtypes are commonly observed in invasive IPMC. Further research is required to establish diagnostic criteria for each subtype so that AC can be offered to those who may benefit most from it. Matrix metalloproteinases (MMPs), which are enzymes that specifically degrade structural constituents of the extracellular matrix have been considered to be associated with tumor progression or poor prognosis. Akashi et al. reported that higher MMP scores correlated with worse prognosis and the tubular type showed higher MMP scores than the colloid type (29).
Patients who received S-1 in our study had less grade 3 or higher adverse events than those in the JASPAC 01 trial (14). None of the three patients who received gemcitabine monotherapy had grade 3 or higher adverse events. AC therefore appears to be well-tolerated, even after pancreatectomy for invasive IPMC.
Our study has several limitations. First, it was a single-center retrospective study; selection biases were inevitable. Second, the small sample size precluded multivariate analyses. As invasive IPMC is less common than PDAC, larger, multicenter studies may be desirable in the future.
In conclusion, AC may benefit selected patients after surgery for invasive IPMC, including those with poorly differentiated adenocarcinoma and those with lymph node metastases.
Footnotes
Authors’ Contributions
Takaaki Furukawa: Conceptualization, Methodology, Data analysis, Writing – Original draft preparation; Manabu Takamatsu: Data analysis, Writing – Reviewing; Yosuke Inoue: Resources, Writing – Reviewing, Takeshi Okamoto: Writing -Reviewing and Editing; Akiyoshi Kasuga, Masato Matsuyama, Takashi Sasaki, Takafumi Mie, Yuto Yamada, Tsuyoshi Takeda, Masato Ozaka: Writing – Reviewing. Atsushi Oba, Hiromichi Ito, Yoshihiro Ono, Takafumi Sato: Resources, Writing – Reviewing; Yu Takahashi, AKIO SAIURA: Resources, Writing – Reviewing and Editing; Naoki Sasahira: Conceptualization, Methodology, Writing – Reviewing and Editing, Supervision.
Supplementary Material
Available at: <https://jfcr-my.sharepoint.com/:w:/g/personal/takaaki_furukawa_jfcr_or_jp/Eft4kVDtSplHixeRX3sjqYQBZoeuqVQ-87HDBcqzUHn3jA?e=WFv2ZV>
Conflicts of Interest
Naoki Sasahira received a research grant from Taiho Pharmaceutical Co., Ltd (Tokyo, Japan). The other Authors declare that they have no conflict of interest regarding this study.
- Received January 8, 2022.
- Revision received April 4, 2022.
- Accepted April 5, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.